cs.CV（2025-08-27）

📊 共 31 篇论文 | 🔗 8 篇有代码

🎯 兴趣领域导航

支柱三：空间感知与语义 (Perception & Semantics) (9 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (9 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (9 🔗2) 支柱一：机器人控制 (Robot Control) (2) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱三：空间感知与语义 (Perception & Semantics) (9 篇)

#题目一句话要点标签🔗⭐
1 MAPo : Motion-Aware Partitioning of Deformable 3D Gaussian Splatting for High-Fidelity Dynamic Scene Reconstruction 提出MAPo以解决动态场景重建中的模糊渲染问题 3D gaussian splatting 3DGS gaussian splatting
2 LabelGS: Label-Aware 3D Gaussian Splatting for 3D Scene Segmentation 提出LabelGS以解决3D场景分割能力不足问题 3D gaussian splatting 3DGS gaussian splatting ✅
3 Seam360GS: Seamless 360° Gaussian Splatting from Real-World Omnidirectional Images 提出Seam360GS以解决360度图像渲染不完美问题 3D gaussian splatting gaussian splatting splatting
4 OpenM3D: Open Vocabulary Multi-view Indoor 3D Object Detection without Human Annotations 提出OpenM3D以解决无人工注释的多视角室内3D物体检测问题 open-vocabulary open vocabulary
5 Scalable Object Detection in the Car Interior With Vision Foundation Models 提出ODAL框架以解决车内物体检测与定位问题 scene understanding foundation model
6 FastAvatar: Towards Unified Fast High-Fidelity 3D Avatar Reconstruction with Large Gaussian Reconstruction Transformers 提出FastAvatar以解决高时间复杂度和数据利用率低的问题 3D gaussian splatting 3DGS gaussian splatting
7 Context-aware Sparse Spatiotemporal Learning for Event-based Vision 提出上下文感知稀疏时空学习以解决事件视觉处理问题 optical flow spatiotemporal
8 Q-Align: Alleviating Attention Leakage in Zero-Shot Appearance Transfer via Query-Query Alignment 提出Q-Align以解决零样本外观转移中的注意力泄漏问题 semantic mapping semantic map structure preservation
9 AutoQ-VIS: Improving Unsupervised Video Instance Segmentation via Automatic Quality Assessment 提出AutoQ-VIS以解决无监督视频实例分割中的质量评估问题 optical flow ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
10 How Multimodal LLMs Solve Image Tasks: A Lens on Visual Grounding, Task Reasoning, and Answer Decoding 提出多模态LLM分析框架以揭示视觉任务处理机制 large language model multimodal visual grounding
11 Grounding Multimodal Large Language Models with Quantitative Skin Attributes: A Retrieval Study 结合定量皮肤属性的多模态大语言模型以提升皮肤病诊断解释性 large language model multimodal
12 CVBench: Evaluating Cross-Video Synergies for Complex Multimodal Understanding and Reasoning 提出CVBench以解决多视频关系推理评估问题 large language model multimodal chain-of-thought ✅
13 AudioStory: Generating Long-Form Narrative Audio with Large Language Models 提出AudioStory以解决长篇叙事音频生成问题 large language model instruction following ✅
14 Multimodal Conditional MeshGAN for Personalized Aneurysm Growth Prediction 提出MCMeshGAN以解决个性化动脉瘤生长预测问题 multimodal ✅
15 AIM: Adaptive Intra-Network Modulation for Balanced Multimodal Learning 提出自适应网络内调制以解决多模态学习不平衡问题 multimodal
16 Mind the Third Eye! Benchmarking Privacy Awareness in MLLM-powered Smartphone Agents 提出大规模基准以评估智能手机代理的隐私意识 large language model multimodal ✅
17 Integrating SAM Supervision for 3D Weakly Supervised Point Cloud Segmentation 提出结合SAM监督以解决3D弱监督点云分割问题 foundation model
18 The Return of Structural Handwritten Mathematical Expression Recognition 提出结构化手写数学表达式识别方法以解决符号对齐问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
19 Discrete Diffusion VLA: Bringing Discrete Diffusion to Action Decoding in Vision-Language-Action Policies 提出离散扩散VLA以解决视觉-语言-动作模型的统一性问题 transformer policy vision-language-action VLA ✅
20 Mitigating Hallucinations in Multimodal LLMs via Object-aware Preference Optimization 提出CHAIR-DPO以减少多模态大语言模型的幻觉问题 DPO direct preference optimization large language model ✅
21 Plug-in Feedback Self-adaptive Attention in CLIP for Training-free Open-Vocabulary Segmentation 提出反馈自适应注意力机制以解决CLIP的开放词汇分割问题 MAE open-vocabulary open vocabulary
22 Multimodal Prototype Alignment for Semi-supervised Pathology Image Segmentation 提出MPAMatch以解决病理图像分割中的模糊边界问题 contrastive learning foundation model multimodal
23 MotionFlux: Efficient Text-Guided Motion Generation through Rectified Flow Matching and Preference Alignment 提出MotionFlux以解决文本驱动运动生成的效率与精度问题 flow matching motion generation
24 Bridging Domain Gaps for Fine-Grained Moth Classification Through Expert-Informed Adaptation and Foundation Model Priors 提出轻量级分类方法以解决蛾类细粒度识别问题 distillation foundation model
25 CODA: Coordinating the Cerebrum and Cerebellum for a Dual-Brain Computer Use Agent with Decoupled Reinforcement Learning 提出CODA框架以解决科学计算中的自主代理执行问题 reinforcement learning generalist agent
26 ATMS-KD: Adaptive Temperature and Mixed Sample Knowledge Distillation for a Lightweight Residual CNN in Agricultural Embedded Systems 提出ATMS-KD框架以提升农业嵌入式系统中的轻量级CNN性能 distillation
27 Self-supervised structured object representation learning 提出自监督结构化物体表示学习以提升视觉理解能力 representation learning

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
28 Ego-centric Predictive Model Conditioned on Hand Trajectories 提出统一的预测模型以解决人机交互中的动作与视觉结果建模问题 manipulation predictive model human-object interaction
29 Improving Generalization in Deepfake Detection with Face Foundation Models and Metric Learning 提出基于人脸基础模型与度量学习的深伪检测框架以提升泛化能力 manipulation foundation model

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
30 Interact-Custom: Customized Human Object Interaction Image Generation 提出Interact-Custom以解决人机交互图像生成中的身份与交互控制问题 human-object interaction HOI CHOIS

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
31 SAT: Supervisor Regularization and Animation Augmentation for Two-process Monocular Texture 3D Human Reconstruction 提出SAT框架以解决单目纹理3D人类重建中的几何模糊问题 SMPL

⬅️ 返回 cs.CV 首页 · 🏠 返回主页